基于力场活动轮廓模型的舌图像分割研究  

Research on tongue image segmentation based on active contour model of force field

作　　者：颜建军[1] 徐姿 郭睿 蒯智杰 王忆勤[3] YAN Jian-jun;XU Zi;GUO Rui;KUAI Zhi-jie;WANG Yi-qin(School of Mechanical and Power Engineering,East China University of Science and Technology,Shanghai 200237,China;Institute of Interdisciplinary Research Complex,Shanghai University of Traditional Chinese Medicine,Shanghai 201203,China;Laboratory of Information Access and Synthesis of Traditional Chinese Medicine Four Diagnosis,Shanghai University of Traditional Chinese Medicine,Shanghai 201203,China)

机构地区：[1]华东理工大学机械与动力工程学院,上海200237 [2]上海中医药大学交叉科学研究院,上海201203 [3]上海中医药大学四诊信息综合实验室,上海201203

出　　处：《中华中医药杂志》2019年第8期3725-3727,共3页China Journal of Traditional Chinese Medicine and Pharmacy

基　　金：国家自然科学基金项目(No.81270050,No.81302913,No.30901897,No.81173199)~~

摘　　要：目的:针对现有舌图像分割方法复杂耗时的问题,文章提出一种基于力场活动轮廓模型的舌图像分割方法。方法:基于轮廓点之间的拉力、斥力以及来自边缘点的吸引力,建立力场活动轮廓模型,进行舌图像的分割。结果:提出的舌图像分割方法不仅能够较好地降低图像内噪声点对轮廓变形的影响,而且能够较准确地提取舌体轮廓,取得了较为满意的舌体分割结果。结论:基于力场活动轮廓模型的舌图像分割方法较好地进行舌图像的分割,对舌诊客观化具有一定实用价值和借鉴意义。Objective: Aiming at the problem that the existing tongue image segmentation method is complicated and time-consuming, this paper proposes a segmentation method of tongue images based active contour model of force field. Methods:Based on the tension and repulsive force between the two contour points, and the attractive force between a contour point and an edge point, the active contour model of force field is established to segment the tongue images. Results: The segmentation method of tongue images proposed in this paper can not only reduce the influence of noise points on the contour deformation, but also obtain satisfactory results of tongue segmentation. Conclusion: The tongue segmentation method can effectively and accurately segment the tongue images, which can provide reference significance for the research of tongue diagnosis objectification.

关 键 词：舌图像分割 活动轮廓模型 轮廓变形 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]